Lactobacillus pentosus MJM60383 Inhibits Lipid Accumulation in Caenorhabditis elegans Induced by Enterobacter cloacae and Glucose

Gut microbiota are known to play an important role in obesity. Enterobacter cloacae, a Gram-negative bacterium, has been considered a pathogenic bacterium related to obesity in the gut. In this study, we established an obesity model of C. elegans by feeding E. cloacae combined with a high glucose diet (HGD), which significantly induced lipid accumulation. An anti-lipid mechanism study revealed that the fatty acid composition and the expression level of fat metabolism-related genes were altered by feeding E. cloacae to C. elegans under HGD conditions. Lactic acid bacteria that showed antagonistic activity against E. cloacae were used to screen anti-obesity candidates in this model. Among them, L. pentosus MJM60383 (MJM60383) showed good antagonistic activity. C. eleans fed with MJM60383 significantly reduced lipid accumulation and triglyceride content. The ratio of C18:1Δ9/C18:0 was also changed in C. elegans by feeding MJM60383. In addition, the expression level of genes related to fatty acid synthesis was significantly decreased and the genes related to fatty acid β-oxidation were up-regulated by feeding MJM60383. Moreover, MJM60383 also exhibited a high adhesive ability to Caco-2 cells and colonized the gut of C. elegans. Thus, L. pentosus MJM60383 can be a promising candidate for anti-obesity probiotics. To the best of our knowledge, this is the first report that uses E. cloacae combined with a high-glucose diet to study the interactions between individual pathogens and probiotics in C. elegans.


Introduction
Obesity can cause many diseases, such as type 2 diabetes, hypertension, cardiovascular disease, and arteriosclerosis due to a combination of genetic factors, fat metabolism disorders, dietary habits, and cultural factors. Consumption of a high-fat diet (HFD) is an important factor leading to obesity. Long-term intake of HFD can lead to the accumulation of fat in animals [1]. Various dietary plans, medications, and surgical procedures have been developed to control obesity. However, there is a need to develop safe and effective anti-obesity therapies to help weight management.
It was discovered that gut microbiota are associated with obesity [2]. Compositions of gut microbiota are different between obese and lean humans. Several studies have shown a decrease in Bacteroidetes but an increase in Firmicutes in obese people [3], suggesting that the HFD diet can influence gut microbial profile and diversity [4]. Interestingly, Enterobacter cloacae was recently discovered to be rich in obese humans [5]. It has been reported that Enterobacter cloacae B29 in combination with a high-fat diet, but not B29 alone, can cause obesity in germ-free mice, suggesting a pathogenic role of B29 in fat metabolism [5]. Enterobacter cloacae can produce lipopolysaccharide (LPS) that can stimulate inflammation in

E. cloacae Combined with a High Glucose Diet-Induced Lipid Accumulation in C. elegans
To establish an obesity model, fat accumulation was measured in the worms by feeding them with E. cloacae in the absence or presence of glucose. As shown in Figure 1A, the intensity of ORO dye was stronger in the HGD-OP50 group and HGD-E. cloacae group than in the ND-OP50 group. The lipid accumulation was increased by 2.1 fold in the HGD-OP50 group and 3.2 fold in HGD-E. cloacae group compared to the ND-OP50 group ( Figure 2B). However, feeding E. cloacae without glucose supplementation didn't increase lipid accumulation. The length and width of the worms in the HGD-E. cloacae group were also longer and wider than the other groups ( Figure 2C,D). ure 2B). However, feeding E. cloacae without glucose supplementation didn't increase lipid accumulation. The length and width of the worms in the HGD-E. cloacae group were also longer and wider than the other groups ( Figure 2C,D). . Statistically significant differences between the treated group and control group as analyzed using the one-way ANOVA with Dunnett's multiple comparison test (** p < 0.01, *** p < 0.001, **** p < 0.0001).  Relative expression level of fat metabolism-related genes in HGD-E. cloacae model. Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison test (* p < 0.05, ** p < 0.01, **** p < 0.0001).
We also determined the expression of genes related to fat metabolism in the HGD-OP50 and HGD-E. cloacae group. In the HGD-OP50 group, the gene expression of asc-2 and nhr-49 were significantly reduced compared to the ND-OP50 group, whereas fat6 and fat7 were significantly upregulated in the HGD-OP50 group ( Figure 2C). However, in the HGD-E. cloacae group, the gene expression of fat6, pept-1, tub-1 and fat7 were significantly increased, whereas the acs-2 gene was significantly down-regulated compared to the HGD-OP50 group ( Figure 2D). Relative expression level of fat metabolism-related genes in HGD-E. cloacae model. Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison test (* p < 0.05, ** p < 0.01, **** p < 0.0001).
We also determined the expression of genes related to fat metabolism in the HGD-OP50 and HGD-E. cloacae group. In the HGD-OP50 group, the gene expression of asc-2 and nhr-49 were significantly reduced compared to the ND-OP50 group, whereas fat6 and fat7 were significantly upregulated in the HGD-OP50 group ( Figure 2C). However, in the HGD-E. cloacae group, the gene expression of fat6, pept-1, tub-1 and fat7 were significantly increased, whereas the acs-2 gene was significantly down-regulated compared to the HGD-OP50 group ( Figure 2D).

Antagonistic Activity of LAB
A total of 70 strains isolated from fermented food were tested for their antagonistic activity against E. cloacae. Among them, 30 strains showed antagonistic activity (Table 1). Strains L. pentosus MJM60298, L. plantarum MJM60335, and L. pentosus MJM60383 showed stronger activity than LGG.

Screening of LAB with Anti-Lipid Accumulation Activity by Using HGD-E. cloacae Model
To test whether the anti-lipid accumulation activity of LAB is related to its antagonistic activity against E. cloacae, six strains (MJM60335, MJM60383, MJM60298, MJM60chosenMJM60454, MJM60556) were chosen to test their anti-obesity activity using the HGD-E. cloacae model. As shown in Figure 3A, the synchronized worms were grown on NGM plates seeded with OP50 or E. cloacae in the presence of glucose (100 mM) for 37 h at 20 • C, then they were transferred onto new NGM plates seeded only with LAB in the presence of glucose. After 18 h, the 55h-old adult worms were harvested to quantify their lipid accumulation by Oil Red O. In our study, the worms did not consume LAB from stage L1-L3 larvae worms.
Among the six strains, MJM60383, MJM60298, and MJM60335 reduced lipid accumulation by 42%, 16%, and 1%, respectively ( Figure 3B). However, MJM60556, MJM6,0454, and MJM60494 did not show activity. Therefore, strain MJM60383 was selected for further study due to its high lipid-lowering activity. ence of glucose. After 18 h, the 55h-old adult worms were harvested to quantify their lipid accumulation by Oil Red O. In our study, the worms did not consume LAB from stage L1-L3 larvae worms.

MJM60383 Reduced Fat Deposition and TG Content Only in HGD-E. Cloacae Model
To determine whether the lipo-reducing activity of MJM60383 is related to the E. cloacae diet, the fat accumulation in the HGD-OP50 model was also determined. Oil Red O staining and quantification showed that feeding of MJM60383 significantly decreased lipid accumulation in the HGD-E. cloacae-induced obesity model, but not in the HGD-OP50 model ( Figure 4A,B). The feeding of MJM60383 decreased lipid accumulation by 58% in the HGD-E. cloacae model. However, fat deposition was not significantly changed by the feeding of MJM60383 in the HGD-OP50 model or the ND-OP50 model ( Figure 4B). . Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison test (** p < 0.01, **** p < 0.0001 compared with the HGD-E. cloacae group).

MJM60383 Reduced Fat Deposition and TG Content Only in HGD-E. cloacae Model
To determine whether the lipo-reducing activity of MJM60383 is related to the E. cloacae diet, the fat accumulation in the HGD-OP50 model was also determined. Oil Red O staining and quantification showed that feeding of MJM60383 significantly decreased lipid accumulation in the HGD-E. cloacae-induced obesity model, but not in the HGD-OP50 model ( Figure 4A,B). The feeding of MJM60383 decreased lipid accumulation by 58% in the HGD-E. cloacae model. However, fat deposition was not significantly changed by the feeding of MJM60383 in the HGD-OP50 model or the ND-OP50 model ( Figure 4B).
In addition, the TG content of worms was reduced by 54% in the HGD-E. cloacae-MJM60383 group ( Figure 4C). Moreover, the width of the worms ( Figure 4D), but not the length, ( Figure 4E) was significantly inhibited compared with HGD-E. cloacae group. In addition, the TG content of worms was reduced by 54% in the HGD-E. cloacae-MJM60383 group ( Figure 4C). Moreover, the width of the worms ( Figure 4D), but not the length, ( Figure 4E) was significantly inhibited compared with HGD-E. cloacae group. . Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison test (** p < 0.01, **** p < 0.0001).

MJM60383 Altered Desaturation Index and Gene Expression
The fatty acid composition and the expression of the genes related to fat metabolism were altered in C. elegans fed with E. cloacae combined with a high glucose diet ( Figure  2A,D). Then, the effect of diet change (feeding of MJM60383) on the ratio of C18:1Δ9/C18:0 and the expression of genes related to fat metabolism was investigated. We found that the feeding of MJM60383 can significantly decrease the ratio of C18:1Δ9/C18:0 by 30% compared to the HGD-E. cloacae group (LAB non-feeding control) ( Figure 5A). In addition, in the MJM60383-feeding group, fat6, pept-1, tub-1 and fat7 genes were significantly decreased, whereas acs-2 and nhr-49 genes were significantly increased compared to the HGD-E. cloacae group ( Figure 5B). . Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison test (** p < 0.01, **** p < 0.0001).

MJM60383 Altered Desaturation Index and Gene Expression
The fatty acid composition and the expression of the genes related to fat metabolism were altered in C. elegans fed with E. cloacae combined with a high glucose diet (Figure 2A,D). Then, the effect of diet change (feeding of MJM60383) on the ratio of C18:1∆9/C18:0 and the expression of genes related to fat metabolism was investigated. We found that the feeding of MJM60383 can significantly decrease the ratio of C18:1∆9/C18:0 by 30% compared to the HGD-E. cloacae group (LAB non-feeding control) ( Figure 5A). In addition, in the MJM60383-feeding group, fat6, pept-1, tub-1 and fat7 genes were significantly decreased, whereas acs-2 and nhr-49 genes were significantly increased compared to the HGD-E. cloacae group ( Figure 5B). The relative expression level of fat metabolism-related genes. Data are expressed as the mean ± SD. Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison tests (**** p < 0.0001, compared with the HGD-E. cloacae group).

Adhesion of LAB to the Caco-2 Cell and Bacterial Colonization in the Gut of C. Elegans
To determine the adhesion ability of LAB to the gut cells, LAB was co-cultured with Caco-2 cells and the percent of attached cells was calculated, as shown in Figure 6. Strain MJM60335, MJM60298, ad MJM60383 showed an adherence of 5%, 20%, and 38% to the Caco-2 cell monolayer, respectively ( Figure 6A).
To confirm whether MJM60383 can colonize in the gut of C. elegans, the CFU of bacteria (E. cloacae and MJM60383) was counted. We found that both E. cloacae and MJM60383 can colonize in the gut of nematodes. In addition, the CFU of E. cloacae in the gut of worms was significantly decreased by feeding MJM60383 compared with the control group (Figure 6B). Data are expressed as the mean ± SD. Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison tests (**** p < 0.0001, compared with the HGD-E. cloacae group).

Adhesion of LAB to the Caco-2 Cell and Bacterial Colonization in the Gut of C. elegans
To determine the adhesion ability of LAB to the gut cells, LAB was co-cultured with Caco-2 cells and the percent of attached cells was calculated, as shown in Figure 6. Strain MJM60335, MJM60298, ad MJM60383 showed an adherence of 5%, 20%, and 38% to the Caco-2 cell monolayer, respectively ( Figure 6A).
To confirm whether MJM60383 can colonize in the gut of C. elegans, the CFU of bacteria (E. cloacae and MJM60383) was counted. We found that both E. cloacae and MJM60383 can colonize in the gut of nematodes. In addition, the CFU of E. cloacae in the gut of worms was significantly decreased by feeding MJM60383 compared with the control group ( Figure 6B). Data are expressed as the mean ± SD. Statistically significant differences between the treated group and control group were analyzed using the one-way ANOVA with Dunnett's multiple comparison tests (**** p < 0.0001, compared with the HGD-E. cloacae group).

Adhesion of LAB to the Caco-2 Cell and Bacterial Colonization in the Gut of C. Elegans
To determine the adhesion ability of LAB to the gut cells, LAB was co-cultured with Caco-2 cells and the percent of attached cells was calculated, as shown in Figure 6. Strain MJM60335, MJM60298, ad MJM60383 showed an adherence of 5%, 20%, and 38% to the Caco-2 cell monolayer, respectively ( Figure 6A).
To confirm whether MJM60383 can colonize in the gut of C. elegans, the CFU of bacteria (E. cloacae and MJM60383) was counted. We found that both E. cloacae and MJM60383 can colonize in the gut of nematodes. In addition, the CFU of E. cloacae in the gut of worms was significantly decreased by feeding MJM60383 compared with the control group (Figure 6B).  In the control group, worms were fed with Ecloacaee only for 55 h. In the HGD-E. cloacae model, worms were conditioned with E. cloacae for 37 h, then transferred to an NGM plate seeded with MJM60383 and incubated for 18 h. Bacteria were calculated as Log CFU in each group (containing 2000 worms). **** p < 0.0001, Statistical significance was determined by a two-tailed unpaired Student's t-test between two groups.

Discussion
Gut microbiota have been thought to be a very important factor in obesity development [2]. Recently, the administration of probiotics has been reported to reduce obesity in mice and humans. However, the mechanism has not been elucidated [20]. A recent study has shown that large amounts of Gram-negative bacteria are present in obese people [21]. These large groups of Gram-negative bacteria can produce high amounts of lipopolysaccharides (LPS) and trigger inflammation in humans and mice [22][23][24][25]. Notably, E. cloacae, an LPS-producing Gram-negative bacteria, was isolated from the gut of an obese human and caused obesity in germ-free mice [5]. E. cloacae can weaken the colonic epithelial barrier and cause obesity through an inflammatory response induced in LPS-stimulated macrophages [6,7]. Therefore, E. cloacae is considered to be an obesity-related pathogenic bacteria in the gut. To study the mechanism of the anti-obesity of probiotics, we established an obesity model by feeding C. elegans with E. cloacae and tested the anti-obesity effect of lactobacilli by establishing an induced obesity model.
Currently, an obesity model in C. elegans was established by adding 100 mM of glucose to the NGM medium. Peng H, et al. reported that lipid accumulation was increased by 30% in the HGD group compared to the normal-diet group [24]. In addition, in their study, gene fat6 and fat7 were significantly increased the in HGD model. However, genes nhr-49 (Nuclear Hormone Receptor NHR-49) and acs-2 were significantly downregulated [25]. Genes fat6 and fat7 can use stearic acid(C18:0) as a substrate to convert to oleic acid (C18:1∆9). The gene nhr-49 encodes a transcription factor involved in the control of pathways to regulate fat consumption and maintain the normal balance of fatty acid saturation by modulating the expression of genes involved in fatty acid beta-oxidation. The deletion of gene nhr-49 increased fat accumulation in C. elegans [13]. Acs-2, one of the downstream genes of nhr-49, is an acyl-CoA synthetase catalyzing the conversion of fatty acid to acyl-CoA which is involved in fatty acid β-oxidation, resulting in a reduction of fatty acids [26]. The gene expression in the HGD model is consistent with our results ( Figure 2C).
To understand the effect of LAB on E. cloacae-related obesity, an obesity model was established by using E. cloacae combined with a high-glucose diet (HGD-E. cloacae). In this model, worms were fed with E. cloacae instead of E. coli OP50, the standard diet for C. elegans. As shown in Figure 1, fat accumulation in the HGD-E. cloacae model was significantly increased compared to the HGD-OP50 group. Moreover, the length and width of worms were significantly increased in the HGD-E. cloacae model. However, worms fed with E. cloacae without glucose supplementation did not show an increase in fat accumulation. The results of C. elegans are consistent with that in mice reported by Fei, N., Zhao et. al [5]. They reported that E. cloacae can produce endotoxin under high-fat diet (HFD) conditions to induce inflammation and contribute to obesity in germ-free mice. However, E. cloacae can't produce endotoxin under normal conditions, so it did not induce obesity. The high-fat diet was also reported to impact bacterial metabolism. Yoshimoto Et Al. reported that HFD altered gut microbiota, leading to an increase in deoxycholic acid, a gut bacterial metabolite known to cause DNA damage [27]. Similarly, Schwiertz et al. found significant differences in SCFA concentrations between lean and obese individuals [28]. Thus, we hypothesized that endotoxin producer E. cloacae could induce obesity in C. elegans under HGD conditions. In addition, we observed that some milky viscous substances were secreted by E. cloacae on the NGM plate (supplied with glucose). We suspected that these substances (water-soluble) might contribute to obesity induction. However, there are no reports indicating that the secretion of E. cloacae that can induce obesity, and whether these substances induce obesity needs further study.
Gut microbiota play a role in the modulation of genes associated with fat storage; specific diets have different capacities to regulate fat metabolism. This includes their micronutrient content, which can impact an organism. Micronutrients are vitamins and trace elements used as metabolites or cofactors in metabolic pathways [29]. In the previous study, genes nhr-49 and acs-2 were significantly downregulated in HGD-induced worms, which is consistent with our results. After feeding E. cloacae to C. elegans, the gene expression of acs-2 was significantly inhibited compared to the HGD-OP50 group. Moreover, the genes fat 6, fat 7, pept-1, and tub-1 were significantly upregulated compared to the HGD-OP50 group. Genes fat 6 and fat7 encode stearoyl-CoA desaturase (SCD), which is known to convert stearic acid (C18:0) to oleic acid and control the relative abundance of saturated and monounsaturated fatty acids in C. elegans [11]. These results suggested that the expression of fat metabolism-related genes is different between the HGD-OP50 and HGD-E. cloacae group. This result indicates that E. cloacae has different capacities to regulate fat metabolism from the E. coli OP50. In the HGD-E. cloacae group, the proportion of fatty acids (from C18 to C20) and the ratio of C18:1∆9/C18:0 were significantly increased compared with the HGD-OP50 group. These results suggested that feeding E. cloacae to C. elegans under HGD conditions might increase fatty acid synthesis in C. elegans by regulating the expression level of the fat6 and fat7 genes.
Tub-1, the TUBBY homolog, functions with kat-1, a 3-ketoacyl-CoA thiolase, to modulate fat metabolism, which is linked to fatty acid β-oxidation [30][31][32]. In addition, RAB-7 (Ras-related protein Rab-7a homolog), a mediated endocytic pathway, might be a target for tubby to regulate fat storage [30][31][32]. In the present study, the expression of the tub-1 gene was increased in the HGD-E. cloacae model, suggesting a possible role of E. cloacae in fatty acid β-oxidation and the endocytic pathway. Along with tub-1, nhr-49, a functional homolog of PPARα, can also regulate β-oxidation [13]. However, the current results showed that E. cloacae did not influence nhr-49, whereas a downstream gene acs-2 was significantly down-regulated by E. cloacae. Although both nhr-49 and tub-1 can modulate fatty acid β-oxidation, E. cloacae can modulate tub-1 but not nhr-49. Further studies are needed to determine the role of E. cloacae in tub-1-mediated fatty acid β-oxidation.
PEPT-1, an intestinal peptide transporter, can regulate fat deposition in the intestine of worms [33]. Previous studies have elucidated that PEPT-1 affected fatty acid absorption by interacting with a sodium-proton exchanger NHX-2 to sustain the intracellular pH of intestinal epithelial cells [34]. In our study, the gene expression of PEPT-1 was significantly increased in the HGD-E. cloacae group. An E. cloacae diet might modify the intracellular microenvironment. Taken together, these results suggest that E. cloacae combined with a high glucose diet might have induced fat accumulation in mainly upregulated fat metabolismrelated genes (fat6, fat7, acs-2). Based on these results, an obesity model feeding E. cloacae to C. elegans under HGD conditions was established. To the best of our knowledge, this is the first report that E. cloacae combined with high glucose can induce fat accumulation in C. elegans.
To screen the lactobacilli with anti-lipid accumulation activity by using HGD-E. cloacaea model, we first tested the antimicrobial activity of LAB against E. cloacae ( Table 2). The strains L. pentosus MJM60383, L. pentosus MJM60298, and L. plantarum MJM60335 showed strong antagonistic activity, which is higher than LGG. Other strains showed moderate activity against E. cloacae. Therefore, strains that showed strong or moderate activity were selected to test their anti-lipid accumulation activity in an HGD-E. cloacae induced obesity model of C. elegans (Figure 3). Among these tested strains, MJM60383 and MJM60298 significantly reduced lipid accumulation in the HGD-E. cloacae induced-obesity model. Although MJM60335 also showed strong antagonistic activity against E. cloacae, MJM60335 did not show inhibitory activity in HGD-E. cloacae induced obese C. elegans. Both MJM60383 and MJM60298 are identified as L. pentosus by the 16S rRNA sequence, and MJM60383 showed stronger activity than MJM60298, which indicated that this activity was strainspecific and not species-specific. Strain MJM60383 only can reduce lipid accumulation in an HGD-E. cloacae inducedobesity model instead of an ND-OP50 and HGD-OP50 model ( Figure 4B). However, the lipid-reducing mechanism of L. pentosus MJM60383 under the HGD-E. cloacae model should be studied further. To explore the difference in inhibitory activity between MJM60383 and other LAB with high antagonistic activity, the adhesion ability of LAB to Caco-2 cells was investigated due to major fat stored in intestinal and skin-like epidermal cells in C. elegans. Caco-2 cell was derived from human colon tissue and has been extensively used as a model for the study of probiotics to colonize the intestinal tract and compete with intestinal pathogens for attachment sites in the gut [35]. MJM60383 showed superior adherent activity to Caco-2 cells compared with MJM60298 and MJM60335. This result demonstrated that MJM60383 might have a high possibility of colonization in the gut of worms. It is necessary to consider comprehensive factors when selecting a probiotic candidate sin. We found that MJM60383 can colonize the gut of worms and reduce the CFU of E. cloacae in the gut of C. elegans. This result demonstrated that MJM60383 can compete with E. cloacae for the binding sites, leading to the elimination of the f E. cloacae from the gut or inhibition of the growth of E. cloacae, which may contribute to its anti-lipid accumulation activity. However, both Caco-2 cells and the gut of C. elegans are different from the gut of humans; it is necessary to verify the colonization of MJM60383 in the human gut through a clinical study in the future.
The role of MJM60383 on fat accumulation was investigated at the gene level (Figure 5B). Feeding L. pentosus MJM60383 to worms can significantly upregulate the expression of the acs-2 and nhr-49 genes, enhancing fatty acid β-oxidation and reducing lipids. Moreover, feeding MJM60383 to C. elegans can significantly down-regulate the expression of fat6 and fat7 genes, thereby affecting fatty acid synthesis and reducing the synthesis of triglycerides, the main component of lipids. Taken together, the gene expression was downregulated by feeding MJM60383 in the HGD-E. cloacae model, which might be attributed to its colonization of the intestine. MJM60383 colonized in the gut to possibly exclude E. cloacae or inhibit the growth of E. cloacae. The ratio of C18:1∆9/C18:0 was also reduced by feeding MJM60383, which confirmed the inhibition of fat6 and fat7 genes ( Figure 5A).

C. elegans Culture and Bacterial Growth Conditions
Wild-type N2 Bristol Caenorhabditis elegans was obtained from the Caenorhabditis Genetics Center. Worms were fed with Escherichia coli OP50 and raised at 20 • C on nematode growth medium (NGM) agar plates. LAB was cultured in deMan-Rogosa-Sharpe (MRS) media at 37 • C for 24 h under anaerobic conditions. Enterobacter cloacae KACC 11403 was obtained from the Korean Agricultural Culture Collection and grown on nutrient media for 24 h at 37 • C under anaerobic conditions.

Establishment of the Induced-Obesity Model by Feeding E. cloacae Combined with a High-Glucose Diet (HGD)
Gravid nematodes were treated with a bleach solution (1 N NaOH: bleach: H 2 O; 2.5:1:0.5) to break down the worm's bodies and obtain the eggs. Harvested eggs were hatched in sterilized M9 buffer (1M KH 2 PO 4 , 0.5M Na 2 HPO 4 , 4M NaCl, MgSO 4 ·7H 2 O, H 2 O/ L) at 20 • C for 16 h to obtain age-synchronized germ-free larvae worms (L1 larvae) [36]. L1 larvae worms were fed with E. cloacae in the absence or presence of glucose to establish an induced-obesity model. In the ND-OP50 group, synchronized worms were fed with OP50 on an NGM plate without glucose supplementation and incubated at 20 • C for 55 h. In the HGD-OP50 group, synchronized L1 worms were fed with OP50 on an NGM plate containing 100 mM glucose and incubated at 20 • C for 55 h. In the ND-E. cloacae group, synchronized L1 worms were fed with E. cloacae on an NGM plate without glucose supplementation and incubated at 20 • C for 55 h. In the HGD-E. cloacae group, synchronized L1 worms were fed with E. cloacae on an NGM plate supplemented with 100 mM glucose and incubated at 20 • C for 55 h.

Oil Red O Staining and Quantification
Oil Red O staining of C. elegans was performed according to the method described before [37]. Briefly, worms were washed from the plates with phosphate-buffered saline supplied with 0.01% Triton X-100 (PBST). After washing, the pellet of worms was treated with 40% isopropanol and incubated for 3 min at room temperature (RT). These worms were then resuspended in 60% Oil Red O solution (5 mg/mL) and rotated at 30 rpm for 2 h at RT. Worms were imaged with a Ci-L microscope and a color camera (Nikon Co., Japan).
To quantify lipid accumulation in the worms, stained worms were resuspended in M9 buffer containing 0.1% Tween-20, and the worm suspension (10 µL) was pipetted onto slides to count the number of worms. To obtain accurate worm numbers in each 10 µL suspension, we repeated the counting independently three times and took the mean. A total of 2000 stained worms were added to 200 µL isopropanol and incubated at 37 • C for 30 min. The lipid content was determined by measuring the dye extracted from the stained worms at an absorbance of 510 nm by a microtiter plate reader (Tecan, Austria).

Fatty Acids Composition
The fatty acid composition of worms in each group was determined according to the method of [25]. Briefly, C. elegans was washed with M9 buffer to remove diet bacteria. Worm lipids were extracted with 1 mL solvent (2% H 2 SO 4 mixed with 98% methanol) at 80 • C for 2 h. After incubation, 1 mL of n-hexane was added and mixed. The solvent was then separated into two layers (methanol and hexane phases). The upper layer (hexane phase) was evaporated by nitrogen gas and then weighed. Each sample contained at least 5 mg of lipid.
The lipids of worms were analyzed by gas chromatography−mass spectrometry (GC-MS, Agilent, USA) using a DB-WAX column (120 mm × 0.25 mm × 0.25 µm). The oven temperature was programmed as follows: the initial temperature was 80 • C, held for 3 min, programmed to 200 • C at 15 • C/min, and finally maintained at 80 • C for 3 min. The fatty acids composition of C. elegans was identified by comparison of the retention time and mass spectra with the commercial standard of fatty acid. The result of fatty acids is expressed as the weight. The desaturation index was then determined by measuring the ratio of C18:1 ∆9/18:0 [38].

Antagonistic Activity of LAB
An antagonistic assay was performed to determine if lactobacilli isolated from fermented food have antimicrobial activity against E. cloacae KACC 11403 [39]. Briefly, a nutrient soft agar (0.7% agar) was seeded with 0.1% (v/v) E. cloacae (1 × 10 7 CFU/mL) and overlaid on MRS agar. Holes were made on MRS plates using a borer. The overnight culture (10 µL) of lactic acid bacteria was poured into the hole on the MRS plate and incubated at 37 • C for 24 h. Lactobacillus rhamnosus GG (LGG) was used as a control. After incubation, the inhibition zone was measured.

Screening of LAB with Inhibitory Activity for Lipid Accumulation
The egg was synchronized as described above. After hatching, L1 worms were conditioned with E. cloacae for 37 h at 20 • C. The worms were then washed with M9 buffer to completely remove E. cloacae and transferred to a new NGM plate (supplemented with 100 mM glucose. A total of 2000 worms from each group were collected and quantified by lipid quantification as described above. To confirm that C. elegans consume LAB, worms were conditioned with E. cloacae for 37 h and transferred to an NGM blank without any bacterial lawn used as a blank group.

Triglyceride (TG)
55-h-old adult worms fed with E. cloacae or LAB were harvested. The TG content in each group (containing 2000 worms) was determined by using a PicoSens™ Triglyceride Assay Kit (Biomix, Guri, Republic of Korea) according to the manufacturer's instructions. Protein content in the supernatant was determined using a Pierce TM BCA protein Assay kit following the manufacturer's instructions. The triglyceride content was normalized by protein [40]. Relative TG content was calculated with the following formula: Relative TG content (100%) = TG sample /TG model × 100 (1)

RNA Extraction and Quantitative Real-Time PCR
To determine the influence of the bacterial diet on the expression of fat metabolismrelated genes, 7000 adult worms from each group were washed with M9 buffer and centrifuged at 1200 rpm for 2 min. The worm pellet was frozen in liquid nitrogen and then thawed completely in a heating block at 40 • C. for five times to destroy the worm's cuticle. For RNA isolation, the worm pellet was treated with 1 mL of Trizol reagent and rotated for 20 min at RT, then 200 µL of chloroform was added to the worm pellet and kept on ice for 15 min. Samples were centrifuged at 14,000 rpm for 15 min at 4 • C. The aqueous phase was transferred to a new microcentrifuge tube containing 500 µL isopropanol. After incubating on ice for 10 min, the mixture was centrifuged at 14,000 rpm for 15 min at 4 • C. After centrifugation, the supernatant was discarded and 1 mL of 75% cold ethanol was added to the pellet. After centrifuging at 12,000 rpm for 5 min at 4 • C, the supernatant was discarded and the RNA pellet was air-dried and dissolved in 30 µL RNase-free water.
For qRT-PCR quantification, RNA was reverse transcribed into cDNA with a PrimeScript TM RT Reagent kit and gDNA Eraser (TAKARA Bio INC, Shiga, Japan) reagent according to the manufacturer's protocol. A real-time quantitative RT-PCR was performed on a Roche Light Cycler ® 96 System (Roche, Basle, Switzerland) using a Maxime TM PCR PreMix kit (LIliF, Seongnam, Republic of Korea). Primers used for each gene are listed in Table 2. Act-1 was used as an internal control. Quantitative PCR conditions were as follows: one cycle at 95 • C for 2 min, followed by 40 cycles at 95 • C for 10 s, 53 • C for 10 s, and 72 • C for 20 s. The fold change was then calculated using the 2 -∆∆Ct method [41]. All experiments were performed in triplicate.

Adherence of LAB to Caco-2 Cell
The Caco-2 cell line was purchased from the American Type Culture Collection (ATCC). The cells were cultured in Dulbecco's Modified Eagle's minimal essential medium (DMEM; Gibco) supplemented with 10% (v/v) heat-inactivated fetal calf serum (Gibco), 2 mM lglutamine (Sigma),100Uml −1 penicillin and 100 mg ml −1 streptomycin (Sigma) at 37 • C in 5% CO 2 condition. For the adhesion assay, Caco-2 monolayers were seeded at a concentra-tion of 5 × 10 5 cells per well in 12-well plates. Cells were treated with LAB suspension (1 × 10 7 CFU/mL) for 2 h when the cells reached 80% confluence. After incubation, the adherent LAB bacteria and Caco-2 cells were detached by adding 1ml of 0.05% (v/v) Triton-X 100, and the CFU of adherent bacteria was calculated by serial dilution and spreading on MRS agar plates. The percentage of LAB adherence was calculated as follows: Adhesion (%) = (CFU of adherent bacteria/CFU of initial inoculated bacteria) × 100 (2) 4.10. Counting of Bacteria in the Gut of C. elegans The bacterial colonization in C. elegans was determined and described by Daniela Uccelletti et al. [42], with minor modifications. Worms (2000 worms in each group) either fed with E. cloacae under HGD conditions or fed with MJM60383 in the HGD-E. cloacae model were harvested at 56 h at 20°C and then washed three times with M9 buffer containing 20 µg/mL gentamycin to remove surface MJM60383 and E. cloacae. Worm pellets were washed with PBS and cracked with 1 mL of M9 buffer-1% Triton X-100. Worm lysates were then diluted with a PBS buffer and plated on Nutrient agar (for E. cloacae) or MRS agar (for MJM60383) at 37 • C for 48 h, respectively. E. cloacae and L. pentosus MJM60383 colonies were quantified and used to calculate the number of bacteria. This experiment was performed in triplicate.

Statistical Analysis
All measurements were repeated independently in triplicate. Results are expressed as mean ± standard deviation (SD). The data obtained were statistically analyzed using GraphPad Prism version 8. A one-way analysis of variance was used to study the significant difference between means, with a significance level of * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Conclusions
To the best of our knowledge, this is the first report that E. cloacae combined with a high-fat diet can induce lipid accumulation in C. elegans. It provides a strategy to study the interaction between individual pathogens and probiotics, which is more effective and less time-consuming than rodent models. L. pentosus MJM60383 reduced lipid accumulation and triglyceride content and regulated the expression of lipid-related genes in C. elegans fed with E. cloacae combined with a high glucose diet, thus it can be a promising candidate for anti-obesity probiotics.

Data Availability Statement:
The raw data used to support the findings of this study will be made available by the authors, without undue reservation, to any qualified researcher.